4 ms·

Does this work theoretically interplanetary (just asking because for science) ?

arjunnarayan 9y ago
No. Once your latency goes beyond single digit seconds, performance will probably collapse. Too many subsystems would time out. in theory it could be made to work (with terrible performance, and extremely long commit-waits due to having to wait until the remote planets get back to you), but I wouldn't architect a planetary spanning distributed database this way. We probably would have to go back to the drawing board and start from scratch.
- therealmarv 9y ago
  Thanks for the long answer. Much appreciated. The question came into my mind when reading some of graphics and specifications.
- pishpash 9y ago
  You'd need to give up on consistency, because there is no such thing when the time of communication is long compared to interval of events. In the long run, ACID is dead.
state_machine 9y ago
[cockroachdb employee] Short answer: no. Long answer: at their closest earth and mars are about 54m km apart, at the furthest it's over 400, with an average of around 225m km, so theoretical latency is varies between 4 and 24 minutes. CockroachDB uses synchronous replication via raft, and that latency would cause problems as would some other setting like our window sizes and their interaction with timeouts.
- politician 9y ago
  > CockroachDB uses synchronous replication via raft Deep space aside, I wish the announcement just said that! I came back to HN for insight into the paragraph about "multi-active availability... an evolution in high availability from active-active replication". Marketing... sometimes... I tell you what.
  - state_machine 9y ago
    Whoops, sorry about that. If you're looking for more on how it works (rocksdb, raft, distributed transactions across multiple raft groups, etc), you might find the design doc interesting: https://github.com/cockroachdb/cockroach/blob/master/docs/design.md https://github.com/cockroachdb/cockroach/blob/master/docs/de...
phlakaton 9y ago
More practically, I note this from Cockroach's document on "Deploy > Recommended Production Settings": "When replicating across datacenters, it’s recommended to use datacenters on a single continent to ensure performance (inter-continent scenarios will improve in performance soon). Also, to ensure even replication across datacenters, it’s recommended to specify which datacenter each node is in using the --locality flag. If some of your datacenters are much farther apart than others, specifying multiple levels of locality (such as country and region) is recommended." In short, IIUC, even _planetary_ deployment doesn't come for free (yet). Perhaps I'm just not well-enough versed yet in how people deal with globally-distributed databases, but I'd love to see the docs dig into this a bit more: practical limits of cluster deployment, recommended strategies and tools (if any) to replicate data between clusters, etc.
pishpash 9y ago
Distance aside, you don't even have a semblance of simultaneity with multiple reference frames, so not even theoretically possible.

No. Once your latency goes beyond single digit seconds, performance will probably collapse. Too many subsystems would time out. in theory it could be made to work (with terrible performance, and extremely long commit-waits due to having to wait until the remote planets get back to you), but I wouldn't architect a planetary spanning distributed database this way. We probably would have to go back to the drawing board and start from scratch.

therealmarv 9y ago
Thanks for the long answer. Much appreciated. The question came into my mind when reading some of graphics and specifications.
pishpash 9y ago
You'd need to give up on consistency, because there is no such thing when the time of communication is long compared to interval of events. In the long run, ACID is dead.

state_machine 9y ago

[cockroachdb employee] Short answer: no. Long answer: at their closest earth and mars are about 54m km apart, at the furthest it's over 400, with an average of around 225m km, so theoretical latency is varies between 4 and 24 minutes. CockroachDB uses synchronous replication via raft, and that latency would cause problems as would some other setting like our window sizes and their interaction with timeouts.

politician 9y ago
> CockroachDB uses synchronous replication via raft Deep space aside, I wish the announcement just said that! I came back to HN for insight into the paragraph about "multi-active availability... an evolution in high availability from active-active replication". Marketing... sometimes... I tell you what.
- state_machine 9y ago
  Whoops, sorry about that. If you're looking for more on how it works (rocksdb, raft, distributed transactions across multiple raft groups, etc), you might find the design doc interesting: https://github.com/cockroachdb/cockroach/blob/master/docs/design.md https://github.com/cockroachdb/cockroach/blob/master/docs/de...

phlakaton 9y ago

More practically, I note this from Cockroach's document on "Deploy > Recommended Production Settings": "When replicating across datacenters, it’s recommended to use datacenters on a single continent to ensure performance (inter-continent scenarios will improve in performance soon). Also, to ensure even replication across datacenters, it’s recommended to specify which datacenter each node is in using the --locality flag. If some of your datacenters are much farther apart than others, specifying multiple levels of locality (such as country and region) is recommended." In short, IIUC, even _planetary_ deployment doesn't come for free (yet). Perhaps I'm just not well-enough versed yet in how people deal with globally-distributed databases, but I'd love to see the docs dig into this a bit more: practical limits of cluster deployment, recommended strategies and tools (if any) to replicate data between clusters, etc.

pishpash 9y ago

Distance aside, you don't even have a semblance of simultaneity with multiple reference frames, so not even theoretically possible.